Missing Links
Useful Statistics about Minorities' Exposure to Traffic Risks Are Hard to Come By

Researchers seeking to fully understand minority populations' exposure to traffic-related risks are hindered by a lack of consistent and comprehensive data. This is due to a number of shortcomings in current data collection systems. Among the failings are inconsistency across ethnic groups, differences in collection methods, and failure to report ethnic, racial, or socioeconomic data at all.  
 

Generally traffic crash data come from two sources, police accident reports and medical statistics such as death certificates and hospital discharge forms. Up until very recently, police accident reports did not contain data about victims' ethnicity, but that should be changing with the recent introduction of a new reporting form that contains fields where such information can be noted.
 

But the two sources-law enforcement and public health-track victims separately. For example, when a person is killed in a traffic accident, the death is recorded on the site as a traffic-related fatality in a police report. That will contain data about how the crash occurred and other technical information at the scene that could contribute to understanding the causes of the crash. But as soon as the scene is cleared, the data about the victim enters the public health realm. Here, information such as ethnicity, type of injury, and specific cause of death is collected, but it is never linked back to the traffic report. To do so requires another level of analysis, which is rarely done because it is difficult and expensive.
 

What data we do have indicate that there is a strong correlation between socioeconomic status and traffic safety risks. Many studies show that significant portions of minority populations-African-Americans, Latinos, and Native Americans in particular-lag in rates of seat belt use, are more at risk for impaired driving, and have higher instances of pedestrian injury and death.
 

Minority persons also experience disproportionately high rates of poverty compared to the non-Latino white population. But because data sources often do not include information about victims' socioeconomic status, such as education level or income, it is difficult to determine the role that socioeconomic status plays, compared to race and ethnicity. This could be useful to know, for example, because individuals with low incomes tend to live where the traffic infrastructure-roads, sidewalks, signals-is deteriorated or lacks many safety measures, as in the case of low-income housing located near busy truck routes and highways, and rural areas where there are inadequate sidewalks or traffic signals. At the same time, car ownership tends to be closely correlated with income, so that poor people tend to drive less, which would lower their exposure to motor vehicle crash risks but raise their exposure to pedestrian-related risks. Being able to determine the importance of ethnicity versus socioeconomic status would help policymakers better tailor interventions to reduce risks.
 

Differences in data collection methods can also prove problematic for researchers because findings can be distorted by the way the information is collected. For example, "self-reported" data, where subjects describe their own behaviors, have produced conclusions quite different from data obtained through government records or observational surveys.
 

Even in a single study, results can be inconsistent when researchers employ more than one approach in gathering data. In a 1999 Accident Analysis and Prevention study on traffic safety among Latino farm workers, data on seat belt use acquired through observation differed significantly from that obtained through interviews. Eighty-six percent of interviewed subjects said they wore lap-shoulder belts, but data gathered through observation of the same farm worker group showed a seat belt use rate of only 37 percent. Researchers also found major discrepancies between interviewee responses and observed data during their analysis of child safety seat use among farm worker groups.
 

Data that describe differences in drinking and driving behavior among ethnic groups also vary with the nature of the study. Latinos, Native Americans, and African-Americans are over-represented in various drinking-driving populations, according to researched based on blood alcohol levels, crash reports, and arrest data. Studies using self-reported data, on the other hand, have concluded that whites have the highest rates of driving after drinking, followed by Latinos, and then African-Americans. (In an instance of inconsistent data collection across ethnic groups, information on Native Americans was not collected in those studies.)
 

While studies yield varying results depending on research methods and researchers' consideration of possibly contributory factors such as socioeconomic status, the number of studies suggesting a strong link between ethnicity and traffic safety is large enough to determine that, as a whole, members of ethnic minorities are more at risk for traffic-related injury or death. The evidence that Latinos, African Americans, and Native Americans in particular are at a disadvantage on the road is strong enough to warrant continued research and prevention measures and public awareness campaigns tailored to these groups.

 

Bibliography

Stiles MC, Grieshop JI. Impacts of culture on driver knowledge and safety device usage among Hispanic farm workers. Accident Analysis and Prevention. 1999; 31: 235-241.

 

---